bug#9329: 24.0.50; `condition-case' with (debug...) is broken

bug#9329: 24.0.50; `condition-case' with (debug...) is broken

[Drew Adams]

(let ((debug-on-error  t)
      (eval-expression-debug-on-error  t))
  (condition-case nil (/ 1 0)
    ((debug error) "Test")))
 
The debugger is not opened.  In Emacs 23.3 it is opened (as the doc says
it should be).
 

In GNU Emacs 24.0.50.1 (i386-mingw-nt5.1.2600)
 of 2011-08-15 on 3249CTO
Windowing system distributor `Microsoft Corp.', version 5.1.2600
configured using `configure --with-gcc (4.5) --no-opt'
 

bug#9329: 24.0.50; `condition-case' with (debug...) is broken

[Lars Magne Ingebrigtsen]

"Drew Adams" <drew.adams@oracle.com> writes:

> (let ((debug-on-error  t)
>       (eval-expression-debug-on-error  t))
>   (condition-case nil (/ 1 0)
>     ((debug error) "Test")))
>
> The debugger is not opened.  In Emacs 23.3 it is opened (as the doc says
> it should be).

---
The car of a handler may be a list of condition names
instead of a single condition name.  Then it handles all of them.
---

So it seems to be doing what it's supposed to be doing.

Is this what you wanted to say instead?

(condition-case error
    (/ 1 0)
  (error
   (debug error)
   "Test"))

-- 
(domestic pets only, the antidote for overdose, milk.)
  bloggy blog http://lars.ingebrigtsen.no/

bug#9329: 24.0.50; `condition-case' with (debug...) is broken

[Drew Adams]

> > (let ((debug-on-error  t)
> >       (eval-expression-debug-on-error  t))
> >   (condition-case nil (/ 1 0)
> >     ((debug error) "Test")))
> >
> > The debugger is not opened.  In Emacs 23.3 it is opened (as 
> > the doc says it should be).
> 
> The car of a handler may be a list of condition names
> instead of a single condition name.  Then it handles all of them.

Yes. So? (debug error) is a list of condition names.  It is the car of a handler
(the only handler).

> So it seems to be doing what it's supposed to be doing.

No.  Check the doc.  Check Emacs 23 - e.g., 23.3.1.

> Is this what you wanted to say instead?
> (condition-case error
>     (/ 1 0)
>   (error (debug error) "Test"))

No.

C-h i, Elisp, g handling errors

"If an error is handled by some `condition-case' form, this
ordinarily prevents the debugger from being run, even if
`debug-on-error' says this error should invoke the debugger.

If you want to be able to debug errors that are caught by a
`condition-case', set the variable `debug-on-signal' to a non-`nil'
value.  You can also specify that a particular handler should let the
debugger run first, by writing `debug' among the conditions, like this:

     (condition-case nil
         (delete-file filename)
       ((debug error) nil))
"

Put the cursor after that sexp.  `C-x C-e'.  Do the same thing in Emacs 23.

bug#9329: 24.0.50; `condition-case' with (debug...) is broken

[Chong Yidong]

"Drew Adams" <drew.adams@oracle.com> writes:

> (let ((debug-on-error  t)
>       (eval-expression-debug-on-error  t))
>   (condition-case nil (/ 1 0)
>     ((debug error) "Test")))
>
> The debugger is not opened.  In Emacs 23.3 it is opened (as the doc says
> it should be).

Thanks.  This regression was due to the 2011-01-26 change to eval.c on
the trunk (r102982).  I've committed a fix.

bug#9329: 24.0.50; `condition-case' with (debug...) is broken

[Stefan Monnier]

>> (let ((debug-on-error  t)
>> (eval-expression-debug-on-error  t))
>> (condition-case nil (/ 1 0)
>> ((debug error) "Test")))
>> The debugger is not opened.  In Emacs 23.3 it is opened (as the doc says
>> it should be).
> Thanks.  This regression was due to the 2011-01-26 change to eval.c on
> the trunk (r102982).

Indeed I hate this "feature" and didn't bother to reimplement it.

> I've committed a fix.

Oh well,


        Stefan